Microsoft tool boosts AJAX Web page response

Doloto application performance optimization tool analyzes workloads and performs code-splitting of existing large Web apps

Microsoft is offering a tool to boost the responsiveness of pages in AJAX Web applications, a Microsoft official said Friday.

The company's Doloto tool analyzes AJAX application workloads and performs code-splitting of existing large Web applications, said S. Somasegar, senior vice president of the Microsoft developer division, in his blog.

AJAX Web applications, he said, bring rich user experiences to the Web but often require a lot of downloading of code. "The result is applications that are frustratingly slow to load and taxing to Web servers," Somasegar said.

[ Related: The OpenAjax Alliance is offering assistance with AJAX widget security. | Keep up with app dev issues and trends with InfoWorld's Fatal Exception and Strategic Developer blogs. ]

The Doloto Web page at MSDN DevLabs describes the application performance optimization tool as especially useful for large and complex Web 2.0 applications with a lot of code, such as Bing Maps and Hotmail. The tool is available for download now. A stand-alone client tool, Doloto profiles an application, calculates code coverage, and rewrites JavaScript code.

"Doloto enables applications to initially transfer only the portion of client-side JavaScript code necessary for application initialization," Somasegar said. The rest of the application code is replaced by short stubs. Function code is transferred in the background or on-demand upon first execution, said Somasegar.

"Since code download is interleaved with application execution, users can start interacting with your Web application much sooner without waiting to download code that implements features they're not currently using," he said.

"In our experiments across a number of AJAX applications and network conditions, Doloto reduced the amount of initial downloaded JavaScript code by over 40 percent, resulting in startup often faster by 30 to 40 percent depending on network conditions," Somasegar said.

A correction was made to this story on Sept. 8, 2009.